Method for predicting arrival and departure procedures based on qar data and airport self-weather data

Abstract

This invention discloses a method for predicting arrival and departure procedures based on QAR data and airport self-observed meteorological data. It involves matching historical QAR actual flight trajectories with intelligence trajectories; extracting the date attribute, time information, and corresponding runway meteorological information from the matched QAR data to construct a terminal area procedure prediction model; training the model on the data; and using the trained model to predict usable terminal area procedures, which are then recommended to air traffic controllers for rapid command and control. Compared to existing technologies, this invention's method for predicting arrival and departure procedures based on QAR data and airport self-observed meteorological data utilizes a large amount of historical QAR data, combined with meteorological and intelligence information, to perform historical data matching and machine learning to identify usable and optimal terminal area procedures, which are then recommended to air traffic controllers for rapid control and command.

Description

Technical Field

[0001] This invention relates to the field of aviation, and more particularly to a method for predicting arrival and departure procedures based on QAR data and airport self-observed meteorological data. Background Technology

[0002] Terminal area air traffic control involves terminal area air traffic controllers directing aircraft arrivals and departures based on the airport's arrival and departure routes, flight schedules, weather conditions, navigation facilities, and other conditions. One of the important aspects of control is directing each aircraft to use the appropriate arrival and departure procedures.

[0003] When directing aircraft in the terminal area, air traffic controllers must provide the arrival and departure procedures used in the terminal area. Typically, both ends of an airport runway can accommodate takeoffs and landings, and each runway end has multiple usable arrival and departure procedure paths. When directing aircraft, air traffic controllers must instruct pilots to use the appropriate arrival and departure procedure IDs, and pilots must comply with the controller's instructions.

[0004] Several factors need to be considered when controllers select a terminal area procedure path:

[0005] Firstly: the arrival / departure procedure ID identifier in the aeronautical information data;

[0006] Secondly, operational standards are affected by meteorological conditions. For example, during takeoff, the tailwind must not exceed x knots, the crosswind must not exceed y knots, and the cloud base must not be lower than z feet during landing.

[0007] Thirdly, the availability of entry and exit procedures can be determined by the availability of the navigation station equipment;

[0008] Fourthly: other factors, etc.;

[0009] Therefore, controllers have to consider multiple factors, resulting in lower command efficiency. If accurate terminal area procedural predictions could be made and the prediction results could be pushed to controllers, it would greatly facilitate controllers to quickly make commands.

[0010] There are two major challenges in terminal area program prediction: First, most prediction models use limited datasets, and a large amount of historical data is needed to achieve accurate predictions; second, existing terminal area trajectory prediction technologies have not been able to combine actual weather, intelligence, navigation equipment and restriction information, and the accuracy of predictions based solely on historical trajectories is limited. Summary of the Invention

[0011] The purpose of this invention is to provide a method for predicting the terminal area of ​​a flight by combining a large amount of historical QAR data and historical airport self-observation meteorological data, which solves the above-mentioned problems.

[0012] To achieve the above objectives, the technical solution adopted by this invention is: an approach and departure procedure prediction method based on QAR data and airport self-observed meteorological data, the method flow of which is as follows.

[0013] Step 1: Match the actual flight trajectory and intelligence trajectory based on historical QAR data to find the best matching flight procedure;

[0014] Step 2: Obtain flight information from the QAR data matched by the terminal area program, and obtain the date attribute d corresponding to flight A from the flight information;

[0015] Step 3: Obtain the takeoff or landing time t corresponding to flight A from the flight information, i.e., A(t), and extract the runway meteorological information q corresponding to time t from the historical airport self-observation meteorological data;

[0016] Step 4: Construct a terminal area program prediction model based on the flight procedures in the matched QAR data, the date attribute and time information in the flight information, and the meteorological information in the airport's self-observed meteorological data;

[0017] Step 5: Conduct supervised learning on the terminal area program prediction model to train the model's data;

[0018] Step 6: Using the trained model, calculate future weather forecasts and takeoff and landing dates and times to predict the available terminal area procedures;

[0019] Step 7: The training model will recommend the predicted terminal area procedures to the controller, who can then refer to the prediction results to quickly issue commands.

[0020] As a preferred approach, in step one, a hidden Markov model is used to match the trajectory model, and then the Viterbi algorithm is used to find the best-matching flight procedure.

[0021] As a preferred option, in the hidden Markov model, r i For the flight segments defined in the intelligence data, z i GPS track points

[0022] Define observation probability ;

[0023] Where, r i For the flight segment defined in the intelligence data; z i GPS track points; The distance from the GPS track point at time t defines the great circle distance of the flight segment; For GPS accuracy, set this value to 4.0;

[0024] Define state transition probability ;

[0025] in, The distance is the great circle distance between the GPS track point at time t+1 and the GPS track point at time t, minus the defined flight segment distance; e is a constant 2.71828; β is an estimated median. The formula is as follows:

[0026] ;

[0027] Define the initial state probability .

[0028] As a preferred option, the optimal path trajectory vector is defined as: The Viterbi algorithm is used to find the closest trajectory vector. The specific formula is as follows:

[0029] .

[0030] As a preferred option, in step two, the flight weekday attribute is defined, and the date attribute is mainly divided into two categories: one is holidays and the other is weekdays, namely holidays A (h) and weekdays A (w).

[0031] Preferably, in step three, the runway meteorological information includes wind speed Ws(t), wind direction Wd(t), and cloud base height Cu(t).

[0032] Preferably, in step five, the supervised learning method from Patron is used to train the terminal area program prediction model.

[0033] The input samples for its learning are,

[0034] d=[h,w], where h is assigned the value 0 and w is assigned the value 1; d represents the date, h represents holidays, and w represents weekdays;

[0035] t=[t1,t2,t3,……t24], where t is the time of flight A's departure or landing, which is divided into vectors according to each hour of the day, with values ​​corresponding to the numbers 1-24;

[0036] q = [Ws, Wd, Cu], where Ws is wind speed, Wd is wind direction, and Cu is cloud base height;

[0037] p=[p1,p2,……pn], replaces the flight procedure name in the intelligence with the number pn. This vector is the flight procedure number used in the QAR data for the corresponding date attribute, time, and weather conditions.

[0038] Its learned output samples are,

[0039] ans = f(d, t, q, p).

[0040] Compared with the prior art, the advantages of the present invention are as follows: The present invention is a method for predicting arrival and departure procedures based on QAR data and airport self-observation meteorological data. Based on a large amount of historical QAR data, combined with meteorological, intelligence and other information, historical data matching and machine learning are performed to find the available and optimal terminal area procedures, and the optimal terminal area procedures are recommended to the controllers so that the controllers can use them for rapid control and command. Attached Figure Description

[0041] Figure 1 This is a flowchart of the method of the present invention;

[0042] Figure 2 This is a comparison diagram of the departure procedure on the aeronautical chart and the corresponding GPS trajectory in the intelligence data of this invention;

[0043] Figure 3 This is a structural diagram of the hidden Markov model of the present invention;

[0044] Figure 4 This is a schematic diagram of path matching according to the present invention. Detailed Implementation

[0045] QAR (Quick Access Recorder) refers to an airborne flight data recording device with protective equipment. It can record continuously for up to 600 hours and can collect hundreds of in-flight data points simultaneously, covering most parameters such as aircraft flight control, flight trajectory, and equipment malfunctions.

[0046] Airport-observed meteorological data is measured by ground-based wired meteorological telemetry instruments installed within the airport. These instruments, with their detection units installed near the runways, automatically measure meteorological elements such as wind direction, wind speed, temperature, air pressure, humidity, cloud base height, and runway visual range for different runways.

[0047] Terminal Area Procedures – Civil aviation regulations specifying the horizontal and profile trajectories that aircraft must follow during takeoff and landing at an airport, and indicating the corresponding speed and flight mode requirements.

[0048] Terminal area guidance equipment refers to a type of navigation equipment deployed at airports, including VHF navigation beacons and instrument landing systems (ILS). This equipment can transmit guidance signals to guide aircraft to take off or land along a safe trajectory.

[0049] The invention will be further described below: a method for predicting arrival and departure procedures based on QAR data and airport self-observed meteorological data. (See also...) Figure 1 The method and process are as follows:

[0050] Step 1: Match the actual flight trajectory and intelligence trajectory based on historical QAR data to find the best matching flight procedure;

[0051] Because QAR data is stored at 1-second intervals, various factors during actual flight can cause a mismatch between the actual flight trajectory and the intelligence trajectory. These factors mainly include differences in turning radius due to aircraft performance, and situations where air traffic control directs either direct or detour flights. Figure 2 As shown, the left image contains intelligence data, with the red line indicating the standard departure procedure, named LUVEN-09D. The green trajectory line on the right is the actual flight path of the QAR, with the red line corresponding to the procedure LUVEN-09D.

[0052] Although the actual trajectory differs from the trajectory specified by intelligence, the real QAR data needs to be matched to the corresponding program before further analysis can proceed. Therefore, the first step is data matching to eliminate mismatched QAR data.

[0053] This invention uses a hidden Markov model to match trajectory models, with r i For the flight segments defined in the intelligence data, z i For GPS track points, such as Figure 3 As shown,

[0054] Define observation probability ;

[0055] Where, r i For the flight segment defined in the intelligence data; z i GPS track points; The distance from the GPS track point at time t defines the great circle distance of the flight segment; For GPS accuracy, set this value to 4.0;

[0056] Define state transition probability ;

[0057] in, The distance is the great circle distance between the GPS track point at time t+1 and the GPS track point at time t, minus the defined flight segment distance; e is a constant 2.71828; β is an estimated median. The formula is as follows:

[0058] ;

[0059] Define the initial state probability ,like Figure 4 As shown.

[0060] Then, the Viterbi algorithm is used to find the best-matching flight procedure, as follows:

[0061] Define the optimal path trajectory vector as: The Viterbi algorithm is used to find the closest trajectory vector. The specific formula is as follows:

[0062] .

[0063] Due to the massive amount of QAR data in the industry, with approximately 5 million flights annually, this is a necessary condition for better subsequent supervised learning. This step involves matching the big data of QAR operations in China's civil aviation industry with intelligence data to identify accurate samples from the massive amount of data for use in subsequent supervised learning.

[0064] Step 2: Obtain flight information from the QAR data matched by the terminal area program, and obtain the date attribute d corresponding to flight A from the flight information;

[0065] Since holidays and workdays are important parameters affecting flight scheduling, this invention defines a flight workday attribute and divides the date attribute into two main categories: holidays and workdays, namely, holidays A (h) and workdays A (w).

[0066] Step 3: Obtain the takeoff or landing time t corresponding to flight A from the flight information, i.e., A(t), and extract the runway meteorological information q corresponding to time t from the historical airport self-observation meteorological data; the runway meteorological information includes wind speed Ws(t), wind direction Wd(t), and cloud base height Cu(t).

[0067] Step 4: Construct a terminal area program prediction model based on the flight procedures in the matched QAR data, the date attribute and time information in the flight information, and the meteorological information in the airport's self-observed meteorological data;

[0068] Step 5: Use the supervised learning method in Pathon to perform supervised learning on the terminal area program prediction model to achieve data training of the model.

[0069] The input samples for its learning are,

[0070] d=[h,w], where h is assigned the value 0 and w is assigned the value 1; d represents the date, h represents holidays, and w represents weekdays;

[0071] t=[t1,t2,t3,……t24], where t is the time of flight A's departure or landing, which is divided into vectors according to each hour of the day, with values ​​corresponding to the numbers 1-24;

[0072] q = [Ws, Wd, Cu], where Ws is wind speed, Wd is wind direction, and Cu is cloud base height;

[0073] p=[p1,p2,……pn], replaces the flight procedure name in the intelligence with the number pn. This vector is the flight procedure number used in the QAR data for the corresponding date attribute, time, and weather conditions.

[0074] Its learned output samples are,

[0075] ans = f(d, t, q, p);

[0076] Step 6: Using the trained model, calculate future weather forecasts and takeoff and landing dates and times to predict the available terminal area procedures;

[0077] Steps two through six of this invention, based on the two major factors affecting regulatory operations—time and meteorological conditions—are used for model training and prediction, which better aligns with the requirements and characteristics of actual regulatory operations.

[0078] Step 7: The training model recommends the predicted terminal area procedures to the controller, who can then quickly issue commands based on the prediction results. The purpose of this step is to provide controllers with prediction results in advance, enabling them to mitigate some risks and improve their work efficiency.

[0079] This invention uses the aforementioned method for predicting arrival and departure procedures based on QAR data and airport self-observation meteorological data. By combining a large amount of historical QAR data with meteorological and intelligence information for historical data matching and machine learning, it can identify the available and optimal terminal area procedures and recommend the optimal terminal area procedures to controllers for rapid control and command.

[0080] The foregoing has provided a detailed description of the approach and departure procedure prediction method based on QAR data and airport self-observed meteorological data provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Modifications and improvements to the present invention are possible without exceeding the concept and scope specified in the appended claims. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for predicting arrival and departure procedures based on QAR data and airport self-observed meteorological data, characterized in that: The method flow is as follows: Step 1: Match the actual flight trajectory and intelligence trajectory based on historical QAR data to find the best matching flight procedure; The trajectory model is matched using a hidden Markov model in which r i are the segments defined in the information data and z i are the GPS trajectory points, Define observation probability ; Where, r i For the flight segment defined in the intelligence data; z i GPS track points; The distance from the GPS track point at time t defines the great circle distance of the flight segment; For GPS accuracy, set this value to 4.0; Define state transition probability ; in, The distance is the great circle distance between the GPS track point at time t+1 and the GPS track point at time t, minus the defined flight segment distance; e is a constant 2.71828; β is an estimated median. The formula is as follows: ； Define the initial state probability ; Define the optimal path trajectory vector as: The Viterbi algorithm is used to find the closest trajectory vector. The specific formula is as follows: ； Step 2: Obtain flight information from the QAR data matched by the terminal area program, and obtain the date attribute d corresponding to flight A from the flight information; Step 3: Obtain the takeoff or landing time t corresponding to flight A from the flight information, i.e., A(t), and extract the runway meteorological information q corresponding to time t from the historical airport self-observation meteorological data; Step 4: Construct a terminal area program prediction model based on the flight procedures in the matched QAR data, the date attribute and time information in the flight information, and the meteorological information in the airport's self-observed meteorological data; Step 5: Conduct supervised learning on the terminal area program prediction model to train the model's data; Step 6: Using the trained model, calculate future weather forecasts and takeoff and landing dates and times to predict the available terminal area procedures; Step 7: The training model will recommend the predicted terminal area procedures to the controller, who can then refer to the prediction results to quickly issue commands.

2. The approach and departure procedure prediction method based on QAR data and airport self-observed meteorological data according to claim 1, characterized in that: In step two, the flight weekday attribute is defined, and the date attribute is mainly divided into two categories: one is holidays and the other is weekdays, namely holidays A (h) and weekdays A (w).

3. The approach and departure procedure prediction method based on QAR data and airport self-observed meteorological data according to claim 1, characterized in that: In step three, the runway meteorological information includes wind speed Ws(t), wind direction Wd(t), and cloud base height Cu(t).

4. The approach and departure procedure prediction method based on QAR data and airport self-observed meteorological data according to claim 1, characterized in that: In step five, the supervised learning method in Pathon is used to train the terminal area program prediction model. Its learning input samples are, d=[h,w], where h is assigned the value 0 and w is assigned the value 1; d represents the date, h represents holidays, and w represents weekdays; t=[t1,t2,t3,……t24], where t is the time of flight A's departure or landing, which is divided into vectors according to each hour of the day, with values ​​corresponding to the numbers 1-24; q = [Ws, Wd, Cu], where Ws is the wind speed, Wd is the wind direction, and Cu is the cloud base height; p=[p1,p2,……pn], replaces the flight procedure name in the intelligence with the number pn. This vector is the flight procedure number used in the QAR data for the corresponding date attribute, time, and weather conditions. Its learned output samples are, ans = f(d, t, q, p).

Images